Here’s What Mira Murati from OpenAI Revealed About the Data Used to Train the Sora AI Model
OpenAI is an AI juggernaut that has quickly moved beyond ChatGPT and now offers its AI model in other forms. We’ve seen AI being used to create images and videos that can be created with a simple prompt or text.
But the company has faced questions about its policies and how it manages to obtain the data used to train its AI model. OpenAI has already gone through these allegations via the New York Times, which has sued the AI giant, claiming that its AI model was trained using its articles without their permission.
So when Mira Murat, CTO of OpenAI, was interviewed recently and asked about the data used to train the new Sora model, the executive clearly wasn’t prepared for the grilling that came his way.
The Wall Street Journal asked Murat a bunch of questions, but we all wanted to see if the company would finally reveal its secret sauce for training AI models. The reporter specifically asked Murat what data OpenAI uses to train its new text-to-video AI model, Sora.
Murat was quoted as saying that open has used publicly available information and licensed information to train Sora. He was further questioned about the source of publicly available information, particularly YouTube and Facebook. OpenAI’s CTO reacted with confusion to this question, choosing not to talk about the data OpenAI uses to train its AI models.
And that’s where it’s become a problem for OpenAI or any other AI company. It is ethically important for these companies to disclose their data sources, as the lack of transparency raises questions about the legitimacy and credibility of the responses produced by AI chatbots. Recent events with Google’s own image generator have added to the concerns.
OpenAI has Microsoft in its wings, but episodes like this don’t fill the public with confidence in the future of AI, especially when you have AI models that are slowly but surely being prepared to replace the workforce. There is also a big concern when OpenAI wants to launch an AI-based search engine in the near future.
